P
US9224007B2ActiveUtilityPatentIndex 91

Search engine with privacy protection

Assignee: FOX JOSHUAPriority: Sep 15, 2009Filed: Sep 15, 2009Granted: Dec 29, 2015
Est. expirySep 15, 2029(~3.2 yrs left)· nominal 20-yr term from priority
Inventors:FOX JOSHUAPELTS MICHAELSHAPIRO EITAN
G06F 16/2228G06F 21/6245G06F 16/248G06F 16/9038G06F 21/84G06F 16/2455G06F 16/9535H04L 63/10G06F 21/62G06F 17/30867G06F 17/30991
91
PatentIndex Score
16
Cited by
207
References
20
Claims

Abstract

A search engine system with privacy protection, including a data indexer configured to create an index of data, a search engine configured to search the index of the data in response to a query, and create a search result set including excerpts from the data, and a privacy protector configured to identify at least one data entity within at least one excerpt of the search result set that meets at least one predefined entity extraction criterion, redact the search result set by removing the data entity from the excerpt, and present the redacted search result set on a computer output device.

Claims

exact text as granted — not AI-modified
What is claimed is: 
     
       1. A system for classifying data in accordance with data access reasons, the system comprising:
 a predefined access reasons database having a plurality of predefined access reasons, wherein each reason defines a purpose to which the data will be applied, once accessed; and 
 at least one processor configured for executing instructions to provide:
 a data classifier configured to create a classification index of data by identifying an association between any of said data and at least one of said predefined access reasons, and 
 storing said association in said classification index; and 
 
 a data access request logger configured to require selection and submittal of an access reason from said predefined access reasons database at least when the classification index includes an access reason associated with the underlying data of any excerpt in a search result set resulting from the processing of a query. 
 
     
     
       2. A system according to  claim 1  wherein said predefined access reasons database includes at least one predefined association between any of said predefined access reasons and at least one predefined data entity type, and wherein said predefined association indicates that access is permitted to any of said data that includes a data entity of said data entity type. 
     
     
       3. A system according to  claim 2  wherein said data classifier is configured to identify said association by
 identifying at least one data entity within any of said data, wherein said data entity is of any of said predefined data entity types, and 
 associating any of said data that includes said identified data entity with said access reasons that is associated with said data entity type of said identified data entity. 
 
     
     
       4. A system according to  claim 1  wherein said data classifier is configured to identify said association by determining that said data matches a statistical profile of a sample data set associated with said access reason. 
     
     
       5. A system according to  claim 1  and further comprising:
 a privacy protector configured to eliminate from a search result set resulting from the processing of said query any search result excerpt except where access to said data underlying said search result excerpt is permitted for said selected access reason as indicated by said association in said classification index between said underlying data and said selected access reason. 
 
     
     
       6. A system according to  claim 1  and further comprising:
 a privacy protector configured to
 identify at least one data entity within at least one excerpt of a search result set resulting from the processing of said query, wherein said data entity meets at least one predefined entity extraction criterion, and 
 redact said search result set by removing said data entity from said excerpt unless said data entity is of a data entity type that is associated with said selected access reason as indicated in said predefined access reasons database as permitting access to said data entity type. 
 
 
     
     
       7. A system according to  claim 1  and further comprising:
 a privacy protector configured to eliminate from said search result set any search result excerpt except where access to said data underlying said search result excerpt is permitted for said selected access reason as indicated by said association in said classification index between said underlying data and said selected access reason. 
 
     
     
       8. A system according to  claim 7  wherein said data access request logger is configured to present for selection only those of said access reasons that are associated with said underlying data of said search result set as indicated in said classification index. 
     
     
       9. A system according to  claim 1  and further comprising:
 a privacy protector configured to
 identify at least one data entity within at least one excerpt of a search result set resulting from the processing of a query, wherein said data entity meets at least one predefined entity extraction criterion, and 
 redact said search result set by removing said data entity from said excerpt; and a data access request logger configured to 
 restore said data entity that was previously removed from said excerpt of said search result set if said data entity is of a data entity type that is associated with said selected access reason as indicated in said predefined access reasons database as permitting access to said data entity type. 
 
 
     
     
       10. A method for classifying data in accordance with data access reasons, the method comprising:
 providing a plurality of predefined access reasons, wherein each reason defines a purpose to which the data will be applied, once accessed; and 
 executing instructions with at least one processor of a programmable data processing apparatus, wherein executing the instructions causes the at least one processor to:
 identify an association between a unit of data and at least one of said predefined access reasons, 
 store said association in a classification index, and 
 require selection and submittal of an access reason from said predefined access reasons when the classification index includes an access reason associated with underlying data of an excerpt in a search result set resulting from processing of a query. 
 
 
     
     
       11. A method according to  claim 10  wherein said providing a plurality of predefined access reasons comprises providing at least one predefined association between any of said predefined access reasons and at least one predefined data entity type, wherein said predefined association indicates that access is permitted to any of said data that includes a data entity of said data entity type. 
     
     
       12. A method according to  claim 11  wherein said identifying an association between a unit of data and at least one of said predefined access reasons comprises:
 identifying at least one data entity within any of said data, wherein said data entity is of any of said predefined data entity types; and 
 associating any of said data that includes said identified data entity with said access reasons that is associated with said data entity type of said identified data entity. 
 
     
     
       13. A method according to  claim 10  wherein said identifying an association between a unit of data and at least one of said predefined access reasons comprises determining that said data matches a statistical profile of a sample data set associated with said access reason. 
     
     
       14. A method according to  claim 10  and further comprising:
 eliminating from a search result set resulting from the processing of said query any search result excerpt except where access to said data underlying said search result excerpt is permitted for said selected access reason as indicated by said association in said classification index between said underlying data and said selected access reason. 
 
     
     
       15. A method according to  claim 10  and further comprising:
 identifying at least one data entity within at least one excerpt of a search result set resulting from the processing of said query, wherein said data entity meets at least one predefined entity extraction criterion; and 
 redacting said search result set by removing said data entity from said excerpt unless said data entity is of a data entity type that is associated with said selected access reason as indicated in said predefined access reasons as permitting access to said data entity type. 
 
     
     
       16. A method according to  claim 10  and further comprising:
 eliminating from said search result set any search result excerpt except where access to said data underlying said search result excerpt is permitted for said selected access reason as indicated by said association in said classification index between said underlying data and said selected access reason. 
 
     
     
       17. A method according to  claim 10  and further comprising:
 identifying at least one data entity within at least one excerpt of a search result set resulting from the processing of a query, wherein said data entity meets at least one predefined entity extraction criterion; 
 redacting said search result set by removing said data entity from said excerpt; 
 restore said data entity that was previously removed from said excerpt of said search result set if said data entity is of a data entity type that is associated with said selected access reason as indicated in said predefined access reasons as permitting access to said data entity type. 
 
     
     
       18. The method of  claim 10 , further comprising:
 searching said classification index of data in response to a query, wherein said query is associated with one of said predefined access reasons, wherein said searching is limited to units of data having said predefined access reasons; 
 creating a search result set including excerpts from said data as a result of said searching; 
 identifying at least one data entity within at least one excerpt of said search result set, wherein said data entity meets at least one predefined entity extraction criterion; 
 redacting said search result set by removing said data entity from said excerpt; and 
 presenting said redacted search result set on a computer output device. 
 
     
     
       19. The method of  claim 18 , further comprising:
 creating a set of excerpts from said data independent from the processing of queries by a search engine; 
 identifying at least one data entity within said set of excerpts that meets at least one predefined entity extraction criterion; and 
 pre-redacting said set of excerpts by removing from said set of excerpts said data entity identified in said set of excerpts, 
 wherein said creating a search result set comprises creating said search result set from said pre-redacted set of excerpts in response to said query. 
 
     
     
       20. A computer program product for classifying data in accordance with data access reasons, the computer program product comprising:
 a computer readable storage device; and 
 computer program instructions operative to
 provide a plurality of predefined access reasons, wherein each reason defines a purpose to which the data will be applied, once accessed, 
 identify an association between a unit of data and at least one of said predefined access reasons, 
 store said association in said classification index, 
 require the selection of an access reason from said access reasons in association with a query to be processed regarding said data, and 
 eliminate from a search result set resulting from the processing of said query any search result excerpt except where access to said data underlying said search result excerpt is permitted for said selected access reason as indicated by said association in said classification index between said underlying data and said selected access reason, 
 wherein said program instructions are stored on said computer readable storage device.

Cited by (0)

No later patents cite this yet.

References (0)

No backward citations on record.